Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Recap of the `funding` experiment

feross.org

Post reply on HN
Print view View original

1–10 of 72 posts

Re: Recap of the `funding` experiment

#3

Post by gvkhna »

Thank you for this in depth discussion about sponsoring open source. I think there aren’t great solutions out there and it was definitely worth trying something new.

I remember reading the original post proposing this concept.

Is it possible the system is working fine as-is and has been for decades? Although maintainers do not get great funding, the community is constantly proposing new solutions, improving standards and disseminating breakthroughs.

If companies were all privately building their own stacks, then they would all maintain their IP as a competitive advantage but won’t be able to share common solutions that they could feasibly replicate as well given time and direction. Only by everyone contributing do we have such a great community.

Unfortunately companies that sit on top of open source code don’t derive all of their value from just the code, although it’s definitely a part of it. They provide usually more than just a technology implementation, including support staff, communication strategy and so on that their stack doesn’t provide them.

I wish more maintainers were better funded, it could definitely help but I think also it’s amazing to see what we’ve collectively been able to accomplish so quickly by people contributing for free and permissively for the general good.

Re: Recap of the `funding` experiment

#4

Post by wakeywakeywakey »

Some suggestions that may see future attempts better received:

1. Try it on a repo with more substance. There were many comments about this project being a light wrapper around eslint config. It may be popular in GitHub stars, but intuition tells me the type of dev who installs a dependency to generate a JSON file is not the type to want to pay for anything; they're looking for a quick `npm install free-solution` for everything.

2. Don't focus on ads, specifically. Anecdotally, devs I know are very sensitive to ads/privacy, much more so than to paying money. Despite your good intentions, your ads will involve analytics/tracking, which you ultimately can't control.

3. Your various posts and web profiles eagerly mention how many downloads you have. The numbers are great. But, it comes across as your primary focus. Here's a test you can use to see if you are truly creating value: stop pushing code. When people start begging for updates, I suspect you will have more success with git bounties than ads.

Re: Recap of the `funding` experiment

#5

Post by jchw »

While this experiment was interesting, I think it was doomed from the get go.

One of the problems with advertising is that it works a lot off of a novelty factor. Like, you can come up with a new technique, and for a brief moment in time this new technique of advertising may be effective. Think popup ads, or those annoying newsletter popups, etc etc etc. But then everyone does it, people start ignoring it, and it becomes immediate noise.

The trouble is, why wouldn't anyone `npm install funding`? Soon npm install would have to block the ads just so its own user experience wouldn't be absolutely demolished. This idea simply wouldn't have scaled.

I love open source and I contribute as much as I can, and have definitely been a long time user and supporter, longer than I've worked in software professionally for sure, but along with all of the "Open source is not about you" points that were brought up, there is one point that may hurt but is nonetheless equally true: You are not entitled to make a living off of open source. There is no reason this would be guaranteed to be possible. It would be nice to live in a world like that, but if it doesn't work, it doesn't work. I'm not for solutions that will just deteriorate open source as I think it is much more valuable to society than just a way to make money, to basically all parties involved. I sympathize with people who pour hours of unpaid work into it, but when you donate your time and effort into something, you don't necessarily get anything in return, much like giving gifts to someone you love.

If you need to make a living, you can always sell software, or work for someone else. There's still opportunities to contribute to open source in both of those cases. I would love to see a detailed breakdown is between open source work that's donated versus funded by companies to understand what the open source community really is; I think it would help our understanding about how open source gets funded today.

Of course, though, it is absolutely people's right to do what they wish with their own projects, and I do wish everyone luck in finding a good way to make contributing to open source more sustainable as a source of income.

Re: Recap of the `funding` experiment

#7

Post by danShumway »

I'm in the early stages of building some tools to automatically lock new issues in my own Open Source repos, and to give Patreon supporters automatic access to comment on locked issues. Anyone will be able to read existing Github/Gitlab issues, but only supporters will be able to join the conversation or file new issues. It's loosely based on some ideas that William Gross[0] and a few other sources have proposed.

You're on Patreon, but most of the experiments you've run so far are around traditional sponsorship, ads, and donations. Is there an ethical/logistical reason you've mostly stuck to that path, or is that just the stuff you've happened to try so far?

This is still in the experimental stages for me, but if I built this in a way that other people could use, I'm curious about whether anyone else would be interested in trying it out -- or if the potential to suppress community involvement is part of the reason why developers have steered away from strategies like this.

[0]: http://wgross.net/essays/give-away-your-code-but-never-your-...

Re: Recap of the `funding` experiment

#8

Post by ohazi »

I think there were two problems here.

1. You hit a nerve with the distribution method. Everybody knows that npm is a ticking time bomb for malware delivery. Developers already tend to react poorly when weird things happen to their build system, and the paranoia around npm made them react explosively.

2. I think targeting developers is probably the wrong approach. If you're serious about tackling this problem, you should find a way target executives, marketing, or developer relations people at companies that use the open-source packages in question, and give them ways to support the project in an official capacity.

For example, give projects a standard way to offer 1-3 top-tier sponsorship spots, and 5-10 second-tier spots. Create an easy way for companies to bid for these spots across a handful of open-source projects that they use and like. The companies that pay $10-20k for a booth at the Open Source Summit conference will absolutely eat this up.

Re: Recap of the `funding` experiment

#9

Post by Ninn »

In my mind this is a problem that must be solved at a packagemanager level, where its both possible to dynamically allocate consumer money based on utilisation/download distrubutions, and other allocation factors, such as more sentimental ones.

In addition i think that just installing a module and then starting to recieve funding is a bad carrot, as i can already feel the drama of people who will refuse to give up abandoned packages and namespaces, just to claim their funding. If you had it a bit more centralised you could atleast punish that kind of behaviour and make people responsible if they want to participate in the ecosystem of payouts.

Re: Recap of the `funding` experiment

#10

Post by bravura »

Thought experiment for funding OSS:

tldr: Your avatar has a glowing halo that shows how much you contribute to the OSS ecosystem. People thus always know how much of an OSS saint you are.

* Instead of gravatar, OSS projects start using fundavatar by default.

* You can contribute $$ to fundavatar to support OSS.

* The more $$ you contribute to fundavatar, the bigger your halo is on your avatar for people to see.

* Fundavatar has a smart voting mechanism to distribute funds among OSS maintainers, that involves your preferences and delegation past the granularity that you don't have preferences. I will describe in more detail if people want to know.

* Crucially, you can get a big halo just through OSS contribution and not lose money by joining fundavatar, because as an OSS maintainer you can pump X% of your revenue back into fundavatar.

Post reply on HN